Ben, ...
My point is we seem to be getting along just fine with ad hoc dissemination of metadata so far, and I don't think we have enough experience of the needs of clients in general to start inventing more formal mechanisms. I notice that plenty of the 'net seems to work just fine without nailing down such questions: e.g. what is the mechanism for disseminating metadata about DNS caching resolvers, as used by almost all clients?
Ad hoc may be working now, but if CT wants to claim that it can scale to accommodate lots of logs and monitors, including self-monitors, then there needs to be a description of how this will work. Not sure I understand your comment about DNS; DHCP is usually used to acquire the address of the nearest DNS server. DHCP was developed because we lacked a standard way to acquire this data and the ad hoc mechanisms didn't scale well. Steve _______________________________________________ Trans mailing list [email protected] https://www.ietf.org/mailman/listinfo/trans
